Beruflich Dokumente
Kultur Dokumente
V. Martin
Knight Piésold, Vancouver, Canada
M. Al-Mamun
SNC Lavalin, Calgary, Canada
A. Small
Klohn Crippen Berger, Fredericton, Canada
ABSTRACT: Understanding the consequences of a tailings dam breach ultimately leads to de-
signing safer dams and properly preparing for emergencies. Guidelines for dam breach studies
are available for water dams, but none of these deal with the hydrodynamic and geotechnical is-
sues related to tailings flows. Since 2013, the Mining Dams Committee of the Canadian Dam
Association (CDA) has been working on developing methodologies to improve the way tailings
dam breach analyses (TDBA) are conducted. Workshops were organized in 2014 and 2015 to
understand the state of practice at the time. In 2016 a CDA Working Group was established to
develop guidelines specific to tailings dams. The Working Group led the development of the
TDBA Bulletin and feedback was obtained on several drafts including a workshop in 2017.
The CDA Technical Bulletin for TDBA will provide the key steps that should be undertaken.
The differences between water retaining and tailings dams will be addressed. The presence of a
supernatant pond and the potential of the tailings to liquefy and flow, are the key parameters in-
fluencing the runout potential and outflow volume. The physical processes occurring during a
TDBA will be discussed with guidance provided on estimating the volume of released tailings
during a breach and predicting where the tailings could flow. The TDBA is planned to be issued
in 2019.
RÉSUMÉ: Pour concevoir des barrages plus sécuritaires et de bien se préparer aux si-
tuations d'urgence, il faut comprendre les conséquences d'une brèche de barrage minier. Des
lignes directrices pour les études des brèches de barrage sont disponibles pour les barrages hy-
drauliques, mais aucun s'appliquer spécifiquement aux problèmes hydrodynamiques et géotech-
nique des écoulements des résidus miniers. Depuis 2013, le Comité des barrages miniers de
l'Association canadienne des barrages (ACB) développe des méthodes pour améliorer comment
les études des brèches de barrages sont menées. Des ateliers ont été organisés en 2014 et 2015
pour comprendre l'état de la pratique à l'époque. En 2016 un groupe de travail de l'ACB était
établi pour développer des lignes directrices spécifiques aux barrages miniers. Le groupe de tra-
vail a développé le bulletin technique et des commentaires sur les brouillons ont été reçus, com-
prenant aussi un atelier en 2017.
Le bulletin technique de l'ACB pour les études des brèches de barrages miniers énonce les
étapes clés à suivre. Les différences entre les barrages hydrauliques et les barrages miniers se-
ront abordées. La présence d'un bassin surnageant et le potentiel de liquéfaction et d'écoulement
des résidus sont les paramètres clés qui influencent le potentiel de ruissellement et le volume de
sortie. Les processus physiques qui se produisent au cours d'une brèche de barrages miniers se-
ront discutés et des conseils sont fournis pour estimer le volume de résidus miniers rejetés lors
d'une brèche et la prévision de l'endroit où les résidus pourraient s'écouler. Le Bulletin sera pu-
blié en 2019.
1 INTRODUCTION
This paper provides an update on a guidance document that is being prepared by the Canadian
Dam Association (CDA) for tailings dam breach analyses (TDBA). The CDA provides a forum
to gather and distill consensus on what constitutes good practice for dam safety in Canada. The
process to develop the Technical Bulletin on Tailings Dam Breach Analyses involved four years
(2016-2019) of collaboration by CDA members. The draft Bulletin is currently undergoing re-
views by various CDA committees and external reviewers.
A Tailings Dam Breach Working Group (WG) was established by the CDA Mining Dams
Committee (MDC) to focus on the development of the guidance for TDBA. Discussions and
workshops were held at annual CDA conferences and a number of revisions were made to in-
corporate the feedback received. The Bulletin is nearing completion and will expand on the
2007 CDA Technical Bulletin Inundation, Consequences and Classification for Dam Safety and
on the 2014 CDA Technical Bulletin Application of Dam Safety Guidelines to Mining Dams.
The CDA Bulletin on TDBA is intended to provide dam safety professionals with guidance
on the general process and scope for conducting these analyses. While the Bulletin will provide
a step by step procedure for such analyses, it is up to the dam owners and the professional engi-
neers to agree on the scope that meets the objectives and the requirements set by the Regulators.
Reliable TDBA and mapping are critical for tailings dam design and safety management as they
help identify and characterize threats to public safety and the environment. The results of the
study are typically presented on inundation and deposition maps (as appropriate) and could be
used for various purposes including dam consequence classification, emergency planning, dam
safety management, failure mitigation planning in case a failure occurred, and mine closure and
dam decommissioning planning.
There is little published guidance specific to TDBA currently available. Practitioners often
refer to guidelines for dam breach analysis of water retaining dams, such as the CDA (2007), or
the Washington State (1992) and FEMA (2013) guidelines. Those guidelines were primarily de-
veloped for water retaining dams, and while all those documents provide details on dam beach
analyses, none of them addresses the hydrodynamic and geotechnical issues related to tailings
flows that are critical to tailings dam breach events. The CDA Bulletin on TDBA will aim to fill
this gap in the literature and will offer a basis for discussion between dam owners, dam safety
professionals, and tailings dam safety regulators.
The physical processes of breaching for tailings dams are complex, as they may include flow
of fluids (supernatant water and eroded and/or fluid tailings), combined with a flow of liquefied
tailings and/or slumping of solid tailings. The rheological behaviour of the released materials
differs from that of water and impacts the total volume of tailings released. Studies of previous
tailings dam failures show that the volume of mobilized tailings could range from as low as 1%
to as high as 100% of the total storage volume (Lucia et al. 1981; ICOLD 2001; Rico et al.
2008; Azam&Li 2010; Small et al. 2017).
A tailings dam failure can be defined as the inability of the dam to meet its design intent,
whether in terms of management, operational, structural or environmental function, resulting in
loss to the stakeholders and the environment. For the purposes of this Bulletin, a tailings dam
failure is a physical breach of the dam followed by an uncontrolled release of stored materials
that could include fluids and tailings. The uncontrolled release of contaminated seepage without
a physical breach of the dam was not considered for dam breach assessment purposes in the
Bulletin.
The characteristics of a tailings impoundment or a tailings storage facility (TSF) and its foun-
dations, construction method, as well as operations, maintenance and environmental conditions
(e.g., rain, wind, earthquake, etc.) inform possible failure modes. Failure modes commonly con-
sidered for tailings dams include collapse and overtopping of the dam, or a combination of these
two modes that are considered under fair weather and flood induced conditions, as required.
Collapse of the dam can occur due to various mechanisms like slope or foundation instability,
piping, erosion, seismic event, etc. Overtopping can occur either due to an extreme flood event,
or inadequate operation of the facility.
This paper provides an overview of the proposed key steps to be undertaken during a TDBA.
Various considerations specific to tailings dams are discussed.
2 KEY STEPS FOR TAILINGS DAM BREACH ANALYSES
In the CDA Bulletin on TDBA, a step by step process will be provided for conducting these
analyses. Figure 1 provides a flow diagram for the different steps to be undertaken in a typical
TDBA. A description of these key steps is provided in subsequent sections, focusing on those
steps that are specific to tailings dams.
Tailings characteristics
2. Background Information and Review
Field and laboratory
investigations
3. Failure Modes and Dam Failure Scenarios
Analysis of flow lique-
faction susceptibility
4. Tailings Dam Breach Assessment Cases
Topographic data
Hydrologic data NO
Supernatant Liquefiable
Pond? Tailings?
YES
Liquefiable
Tailings? YES NO
YES
NO
5. Hydrologic Analysis
6. Breach Analysis
7. Runout Analysis
9. Sensitivity Analysis
It is worth noting that saturated loose contractive tailings materials can liquefy and demon-
strate high flowability when the moisture content is relatively high and the solids content is rela-
tively low. On the other hand, consolidated and densified tailings that would have a relatively
lower moisture content and higher solids content, may not demonstrate the same flowability as
the looser, less consolidated materials. Lowering the moisture content in a given tailings slurry
(liquefied saturated tailings) would change its rheological characteristics from high flowability
(at high moisture contents) to semi-flowable and then to non-flowable (at relevant lower mois-
ture contents). These variations in tailings flowability can be evaluated by considering the rhe-
ology to soil mechanics continuum, as discussed in Adams et al. (2017a, 2017b.), and MEND,
2017.
Examples of the four TDBA cases with corresponding relevant photos shown on Figure 2 are
as follows:
Case 1A – the Merrispruit tailings dam failure occurred in South Africa in 1994 (Figure
2a), or the 1978 Mochikoshi dam failure in Japan.
Case 1B – the Mount Polley TSF failure occurred in British Columbia, Canada, in 2014
(Figure 2b).
Case 2A – the Fundao dam failure in Brazil in 2015 (Figure 2c), the 1994 Tapo Canyon
tailings dam failure in the United States, or the recent Feijão dam failure near Brumadi-
nho, Brazil, in January 2019.
Case 2B – the Clinton Creek Mine failure in Yukon, Canada, in 1985 (Figure 2d), the
1950 Castle Dome failure in the United States, or the 2018 Cadia dam failure in Austral-
ia.
(a) Case 1A: Merrispruit 1994 (b) Case 1B: Mount Polley 2014
(c) Case 2A: Fundao 2015 (d) Case 2B: Clinton Creek 1985
Figure 2. Examples of past failures and tailings dam breach study cases (Photo sources: [a]
http://www.tailings.info/casestudies/merriespruit.htm; [b] http://www.cbc.ca/news/canada/british-
columbia/mount-polley-mine-spill-78-larger-than-1st-estimates-1.2755974; [c]
http://g1.globo.com/minas-gerais/fotos/2015/11/barragem-se-rompe-e-distrito-de-mariana-e-
inundado.html#F1833790; and [d] http://powergeolab.com/fieldsites/)
Table 2. Flow behaviour as a function of solids concentration (adapted from O’Brien 1986)
Sediment Concentration
Dam Breach
Flow Type Flow Characteristics
Study Cases
by Volume by Weight
Water Water flood with conventional suspend-
<0.20 <0.41
Flood ed load and bedload
Distinct wave action; fluid surface; all
0.20 - 0.30 0.41 - 0.54 particles resting on bed in quiescent flu-
id condition
Case 1A Separation of water on surface; waves
Case 1B 0.30 - 0.35 0.54 - 0.59 travel easily; most sand and gravel has
settled out and moves as bedload
Marked settling of gravels and cobbles;
Mud Flood spreading nearly complete on horizontal
0.35 - 0.40 0.59 - 0.65
surface; liquid surface with two fluid
phases appears; waves travel on surface
Flow mixes easily; shows fluid proper-
ties in deformation; spreads on horizon-
tal surface but maintains an inclined flu-
0.40 - 0.45 0.65 - 0.69
id surface; large particle (boulder)
setting; waves appear but dissipate rap-
idly
Case 2A Flow spreading on level surface; cohe-
0.45 - 0.48 0.69 - 0.72
sive flow; some mixing
If there is no supernatant pond present in the TSF or near the breach, the breaching process is
not driven by the discharge of fluids. An initial flood wave (Process I) cannot be formed similar
to Cases 1A and 1B, and the breaching process would be more similar to that described for Pro-
cess II. In Case 2A, the tailings mass has a potential to undergo liquefaction and become flowa-
ble due to various static or cyclic (seismic) trigger mechanisms including lateral unloading re-
sulting from containment removal due to a dam breach. The released tailings mixture would
likely behave as a mudflow, progressively depositing downstream, and may have a solids con-
centration higher than 70% by weight (about 50% by volume), depending on the dry density of
the stored tailings and the volume of interstitial water. Based on observations from past failures,
the post-liquefied residual angle in the TSF would be expected to be shallow at 3.5-5 (6-9%),
while the slope of the tailings deposited downstream may be even shallower at 1-4 (2-7%)
(Lucia et al. 1981, Blight and Fourie, 2003). The slope of the tailings deposited downstream
would also depend on the downstream topography.
In Case 2B, the tailings mass would not have the potential to liquefy and develop flowable
characteristics. The volume of mobilized tailings can be estimated through slope failure mecha-
nisms. The breaching process may be modelled as a landslide, where the materials would not
flow, but block sliding or slow creep deformation could occur in advance of the failure (as
shown in Table 2). The residual angle in the TSF is expected to be much steeper than in Case
2A, and probably closer to the angle of repose for the deposited tailings material.
2.5 Hydrologic analysis
The hydrologic analysis involves determining the starting elevations and volumes for the super-
natant pond in the TSF, and the discharges in the upstream and downstream drainage networks
for the fair weather and flood induced dam breach events. These key hydrologic parameters are
used to determine the breach outflow volumes and to conduct the downstream flood routing.
The maximum normal supernatant pond elevation and the corresponding pond volume are
typically selected for assessing fair weather failure scenarios. For operating dams, it is necessary
to assess whether overtopping is a possible credible scenario based on the past management and
operation of the TSF. If overtopping due to inadequate operations, and insufficient storage vol-
ume and freeboard is considered credible, a pond level coincident with the top of the dam may
need to be selected for the fair weather scenario. The maximum normal pond elevation or the
spillway invert elevation (if applicable) can be selected as the initial supernatant pond elevation
at the onset of a flood event for assessing flood induced failures. It is important to consider the
duration of the inflow design flood (IDF) that was used in the TSF design, as it can have a pro-
found impact on the storm inflow volume, and consequently on the magnitude of the dam
breach outflow volume.
The hydrologic conditions upstream of the TSF that can provide inflows to the TSF may im-
pact the volume of free water in the supernatant pond, and subsequently the volume of mobi-
lized tailings in the breach outflow. For fair weather failures, the runoff volume from the up-
stream drainage network is typically either diverted or included in the normal operational
supernatant pond range. For flood induced failures, however, the flood runoff from the catch-
ment upstream of the TSF needs to be accounted for when determining the additional volume of
free water in the TSF, which can mobilize additional tailings. Diversion structures located up-
stream of the TSF should be assumed to be non-operational for flood induced failures, if they
were designed for smaller return period flood events than the TSF itself.
The assessment of hydrologic conditions downstream of the dam have a profound impact on
the breach flood wave routing, which is similar to the flood routing for a breach of water retain-
ing dams. For fair weather failures, the downstream flows are typically assumed to be equiva-
lent to the mean annual discharge (MAD). For flood induced failures, it is typically assumed
that the storm causing the flood being considered for the breach analysis is centered over the
TSF. Pre-breach flood flows in the drainage network immediately downstream of the facility
should then be equal to the same flood event that was assumed for the breach and then prorated
with distance from the facility depending on the extent of the model.
3 CONCLUSIONS
The Working Group of the CDA’s MDC is nearing completion of a Technical Bulletin specific
to tailings dam breaches. The manuscript is intended to provide dam safety professionals with
guidance on the general process and scope of these types of analyses. Reliable TDBA are critical
for tailings dam design and safety management as they help identify and characterize threats to
public safety and the environment. In this paper, the major differences between breaches of tail-
ings dams and water retaining dams were discussed and the key steps for conducting TDBA
were presented. The presence of a supernatant pond and the potential of the tailings mass to liq-
uefy and flow, are considered to be the key parameters influencing the runout potential and the
outflow volume. The physical processes occurring during a TDBA were discussed, including es-
timating the volume of released tailings materials during a breach.
The results of the TDBA can be used for various purposes including dam consequence classi-
fication, emergency planning, dam safety management, failure mitigation planning, and closure
planning.
4 ACKNOWLEDGEMENTS
The authors would like to extend their gratitude to the CDA and the MDC for the opportunity to
develop this Technical Bulletin and present it at ICOLD 2019. We thank all WG members who
contributed to the development of this Bulletin, all the reviewers for their valuable input, and
the many attendees of various workshops who provided constructive feedback. The WG wel-
comes comments on the proposed guidance document. Please contact the authors if you would
like to discuss any aspect of the proposed guidance document or provide feedback.
5 REFERENCES
Adams, A., Friedman, D., Brouwer, K., and Davidson, S. 2017a. Tailings Impoundment Stabilization to
Mitigate Mudrush Risk. Proceedings of the 85th Annual Meeting of the International Commission on
Large Dams (ICOLD). July 5. Prague, Czech Republic.
Adams, A., Friedman, D., and Davidson, S. 2017b. Characterizing and Stabilizing a Historical Tailings
Facility: The Rheology to Soil Mechanics Continuum. Proceedings of the 2017 Tailings and Mine
Waste Conference. November 5 to 8. Banff, Alberta, Canada.
Azam, S. and Li, Q. 2010. Tailings Dam Failures: A Review of the Last One Hundred Years. Geotech.
News. 28.
Blight, G. E. and Fourie, A. B. 2003. A review of catastrophic flow failures of deposits of mine waste
and municipal refuse. 2003, Proceedings of the international conference FSM2003, Naples: 1–17.
Canadian Dam Association (CDA). 2007. Technical Bulletins. Published as a set to supplemental Dam
Safety Guidelines. Complete listing available on CDA website at www.cda.ca.
Canadian Dam Association (CDA), 2014. Technical Bulletin: Application of Dam Safety Guidelines to
Mining Dams.
Chen, H. and Lee C.F., 2002. “Runout Analysis of Slurry Flows with Bingham Model”. ASCE, Journal of
Geotechnical and Geoenvironmental Engineering. 128: 1032-1042.
Chen, H. and Becker, D. 2014. “Dam Breach Tailings Runout Analysis”. Proceedings of Canadian Dam
Association, 2014 Annual Conference, Banff, Alberta, Oct. 4-9, 2014.
FEMA (Federal Emergency Management Agency), 2013. Federal Guidelines for Inundation Mapping of
Flood Risks Associated with Dam Incidents and Failures. First Editions. FEMA P-946, July 2013.
Federal Energy Regulatory Commission (FERC). 1993. Engineering Guidelines for the Evaluation of
Hydropower Projects, Chapter II: Selecting and accommodating Inflow Design Flood for Dams, Of-
fice of Hydropower Licensing, October 1993, 57 p.
Fread, D.L. 1988 (revised 1991). BREACH: An Erosion Model for Earthen Dam Failures. National
Weather Service, National Oceanic and Atmospheric Administration (NOAA), Silver Spring, Mary-
land.
Fontaine D.D. and V. Martin. 2015. Tailings Mobilization Estimates for Dam Breach Studies. Proceed-
ings of the 2015 Tailings and Mine Waste Conference. Vancouver, BC, October 26-28, 2015.
ICOLD (International Commission on Large Dams). 2001. Tailings Dams – Risk of Dangerous Occur-
rences: Lessons learnt from practical experience. Bulletin 121.
Jeyapalan J. K., Duncan J. M. and Seed H. B. 1983. Analyses of flow failures of mine tailings dams.
ASCE, Journal of Geotechnical Engineering, 109(2); 150-171.
Lucia P. C, Duncan J. M. and Seed H. B. 1981. Summary of research on case histories of flow failures of
mine tailings impoundments. Mine Waste Disposal Technology, Proceedings of the Bureau of Mines
Technology Transfer Workshop, Denver, Colorado, 46-53.
Martin, V., Fontaine, D.D., and J.G. Cathcart. 2015. Practical Tools for Conducting Tailings Dam Breach
Studies. Proceedings of Canadian Dam Association 2015 Annual Conference, Mississagua, ON. Oct.
5-8, 2015.
McDougall, S. and Hungr, O. 2004. “A model for the analysis of rapid landslide motion across three-
dimensional terrain”. Canadian Geotechnical Journal. 41: 1084-1097.
McDougall, S. 2006. A new continuum dynamic model for the analysis of extremely rapid landslide mo-
tion across complex 3D terrain. Ph.D. Thesis, The University of British Columbia.
Mine Environment Neutral Drainage (MEND) Program. 2017. Study of Tailings Management Technolo-
gies. MEND Report 2.50.1, October 2017.
O’Brien, J.S. 1986. Physical Processes, Rheology, and Modeling of Mud Flows. Ph.D. Dissertation, Col-
orado State University, Fort Collins, Colorado.
Rico, M., Benito, G. and Diez-Herrero, A., 2008. “Floods from Tailings Dam Failures”, Journal of Haz-
ard Management. pp. 79-87.
Small, A., James, M. and M. Al-Mamun. 2017. Advancing the State of Practice for Tailings Assessment
Using Empirical Correlations. Proceedings of the Canadian Dam Association 2017 Annual Confer-
ence, Kelowna, BC, Oct. 16-18, 2017.
U.S. Army Corps of Engineers (USACE). 2007. Risk Assessment for Dam Safety, Dam Failure Analysis
Toolbox, September 2007.
Wahl, L. Tony. 1998. Prediction of Embankment Dam Breach Parameters, A Literature Review and
Needs Assessment. Dam Safety Office, Water Resources Research Laboratory, US Department of the
Interior, Bureau of Reclamation, DSO-98-04.
Wahl, L. Tony. 2014. Evaluation of Erodibility-Based Embankment Dam Breach Equations. Technical
Service Center, Hydraulic Investigations and Laboratory Services Group, U.S. Department of the Inte-
rior, Bureau of Reclamation. Denver, Colorado. Hydraulic Laboratory Report HL-2014-20, June 2014.
Walder, J.S. and J.E. O’Connor. 1997. “Methods for Predicting Peak Discharge of Floods Caused by
Failure of Natural and Constructed Earth Dams.” Water Resour. Res., Vol. 33(10), 12.
Wang X., Morgenstern N. R., and Chan D. H., 2010. A model for geotechnical analysis of flow slides and
debris flows, Canadian Geotechnical Journal 47: 1401–1414.
Washington State Department of Ecology. 1992. Dam Safety Guidelines Technical Note 1: Dam Break
Inundation Analysis and Downstream Hazard Classification. Revised October 2007.